With the increasing pervasiveness of digital cameras, the Internet, and social networking, there is a growing need to catalog and analyze large collections of photos and videos. In this dissertation, we explore unconstrained still-image and video-based face recognition in real-world scenarios, e.g. social photo sharing and movie trailers, where people of interest are recognized and all others are ignored. In such a scenario, we must obtain high precision in recognizing the known identities, while accurately rejecting those of no interest. Recent advancements in face recognition research has seen Sparse Representation-based Classification (SRC) advance to the forefront of competing methods. However, its drawbacks, slow speed and sensitivity to variations in pose, illumination, and occlusion, have hindered its wide-spread applicability. The contributions of this dissertation are three-fold: 1. For still-image data, we propose a novel Linearly Approximated Sparse Representation-based Classification (LASRC) algorithm that uses linear regression to perform sample selection for l1-minimization, thus harnessing the speed of least-squares and the robustness of SRC. On our large dataset collected from Facebook, LASRC performs equally to standard SRC with a speedup of 100-250x. 2. For video, applying the popular l1-minimization for face recognition on a frame-by-frame basis is prohibitively expensive computationally, so we propose a new algorithm Mean Sequence SRC (MSSRC) that performs video face recognition using a joint optimization leveraging all of the available video data and employing the knowledge that the face track frames belong to the same individual. Employing MSSRC results in a speedup of 5x on average over SRC on a frame-by-frame basis. 3. Finally, we make the observation that MSSRC sometimes assigns inconsistent identities to the same individual in a scene that could be corrected based on their visual similarity. Therefore, we construct a probabilistic affinity graph combining appearance and co-occurrence similarities to model the relationship between face tracks in a video. Using this relationship graph, we employ random walk analysis to propagate strong class predictions among similar face tracks, while dampening weak predictions. Our method results in a performance gain of 15.8% in average precision over using MSSRC alone.

This thesis deals with copyright protection and content authentication for still images. New blind transform domain block based algorithms using one-level and two-level Discrete Wavelet Transform (DWT) were developed for copyright protection. The mobile number with international code is used as the watermarking data. The robust algorithms used the Low-Low frequency coefficients of the DWT to embed the watermarking information. The watermarking information is embedded in the green channel of the RGB colour image and Y channel of the YCbCr images. The watermarking information is scrambled by using a secret key to increase the security of the algorithms. Due to the small size of the watermarking information comparing to the host image size, the embedding process is repeated several times which resulted in increasing the robustness of the algorithms. Shuffling process is implemented during the multi embedding process in order to avoid spatial correlation between the host image and the watermarking information. The effects of using one-level and two-level of DWT on the robustness and image quality have been studied. The Peak Signal to Noise Ratio (PSNR), the Structural Similarity Index Measure (SSIM) and Normalized Correlation Coefficient (NCC) are used to evaluate the fidelity of the images. Several grey and still colour images are used to test the new robust algorithms. The new algorithms offered better results in the robustness against different attacks such as JPEG compression, scaling, salt and pepper noise, Gaussian noise, filters and other image processing compared to DCT based algorithms. The authenticity of the images were assessed by using a fragile watermarking algorithm by using hash function (MD5) as watermarking information embedded in the spatial domain. The new algorithm showed high sensitivity against any tampering on the watermarked images. The combined fragile and robust watermarking caused minimal distortion to the images. The combined scheme achieved both the copyright protection and content authentication.

This thesis deals with copyright protection and content authentication for still images. New blind transform domain block based algorithms using one-level and two-level Discrete Wavelet Transform (DWT) were developed for copyright protection. The mobile number with international code is used as the watermarking data. The robust algorithms used the Low-Low frequency coefficients of the DWT to embed the watermarking information. The watermarking information is embedded in the green channel of the RGB colour image and Y channel of the YCbCr images. The watermarking information is scrambled by using a secret key to increase the security of the algorithms. Due to the small size of the watermarking information comparing to the host image size, the embedding process is repeated several times which resulted in increasing the robustness of the algorithms. Shuffling process is implemented during the multi embedding process in order to avoid spatial correlation between the host image and the watermarking information. The effects of using one-level and two-level of DWT on the robustness and image quality have been studied. The Peak Signal to Noise Ratio (PSNR), the Structural Similarity Index Measure (SSIM) and Normalized Correlation Coefficient (NCC) are used to evaluate the fidelity of the images. Several grey and still colour images are used to test the new robust algorithms. The new algorithms offered better results in the robustness against different attacks such as JPEG compression, scaling, salt and pepper noise, Gaussian noise, filters and other image processing compared to DCT based algorithms. The authenticity of the images were assessed by using a fragile watermarking algorithm by using hash function (MD5) as watermarking information embedded in the spatial domain. The new algorithm showed high sensitivity against any tampering on the watermarked images. The combined fragile and robust watermarking caused minimal distortion to the images. The combined scheme achieved both the copyright protection and content authentication.

This thesis concerns the background reconstruction from several photographs (so called depopulation scene efect). There are presented methods for obtaining the background from video and discussion of their use for photographs. The greatest emphasis is placed on the Gaussian mixture model and effort to improve this algorithm due to static image. The photographs should be taken with a tripod.

L’immense panoplie d’ouvrages et de travaux consacrés tant au cinéma qu’au jeu vidéo s’évertue très souvent à examiner une œuvre quelle qu’elle soit dans son ensemble, ou à relier des motifs similaires entre plusieurs cas pour en dégager des concepts, des notions et des lignes théoriques. Plus rarement la question des plans, des segments bien spécifiques d’un film ou d’un jeu a été abordée, en particulier les fragments ou les détails se plaçant en porte-à-faux par rapport à la continuité habituelle du reste de l’œuvre. Soudainement, ce personnage d’apparence statique en avant-plan contemple une temporalité affolée en arrière-plan, ou une section de ce film génère de la curiosité vis-à-vis du reste de sa narration. Ces points spéciaux étonnent, surprennent, et attirent l’attention en se demandant quelle est leur fonction aussi bien narrative qu’esthétique. La philosophie antique et ses synthèses contemporaines inspirent un terme, celui de « résistance vitaliste » qui ne frictionne qu’avec une section de la structure sans en chambouler les fondations, dans le seul but de manifester son existence, mue par la volonté de se déclarer au grand jour. Les conséquences sur les protagonistes de l’histoire n’en sont pas moins négligeables, car ces segments spéciaux requièrent également une attention de leur part, et alimentent leur désir de résister à leur tour, tantôt dans l’alignement esthétique de ce point de l’œuvre, tantôt contre. La même philosophie antique désigne dans pareil cas de figure une démarche de délibération, de jugement du pour et du contre, première phase essentielle pour entrer en résistance. Le héros devient alors Sage – également d’inspiration antique –, un être apte à éclairer les zones obscures et à percer l’apparente stabilité de l’autorité en place en s’attaquant à ces nœuds névralgiques. La présente thèse s’accordera à déplier cette analyse sur quatre chapitres qui l’emmèneront du particulier au général, en considérant la matière filmique même et l’entité enregistreuse (la caméra) en première instance, avant de passer aux relations entre la matière du médium et le personnage qui s’en délie, puis aux fragmentations des univers diégétiques en présence, pour enfin questionner le contexte global dans lequel l’œuvre est nécessairement saisie, fournissant les clefs de la résistance par « guérilla ». La délibération, le Sage et la résistance seront constamment au cœur des préoccupations, dès lors que le Diable se cache dans les détails. / The tremendous amount of academic papers dedicated both to cinema and video games often examine any work through its general structure, or link similar motives between several examples in order to conceptualize notions or theoretical frames. The problem of shots, or of specific fragments in a film or a video game, has been rarely questioned, particularly some peculiar details that stick out and go against the overall flow of the work involved. For instance, a static-looking character in the foreground of a picture gazes at another kind of temporality in the background (in slow or fast motion, for example), or a section of a film seems aesthetically odd compared to the rest of the narration it is inserted in. Those special bits tend to surprise the viewer, and attract their attention whilst the latter wonders about both the narrative and aesthetic role of those fragments. The antic philosophy and its modern synthesis inspire the concept of “vitalist resistance”, designating a contradictory element (even a counter-element) in one or a couple of parts of the whole structure without intending to make its foundations collapse, for the sole purpose of manifesting itself, motivated by the desire to exist in plain sight. The consequences on the protagonists of the story are far from being insignificant, since those particular segments call out to their attention as well, and incite them to resist, either alongside the aesthetic involved, either against. In such case, the same philosophy points out the necessity of making decisions, judging the pros and cons as the essential first step to resisting. The hero thus becomes a Sage – inspired by the antic philosophy as well –, in other words, an individual capable of enlightening obscure parts of the world and shaking the apparent stability of the present authority, by mincing onto those delicate bits. This thesis spreads out its analysis over four chapters that will develop the actual subject from the particular to the general, first by considering the film matter itself and the recording entity (the camera), before examining the connections between that matter and the character who unbinds himself from it, the shattering of diegetic universes, and finally the global context in which the work (film or video game) is inevitably set in, leading a “guerilla” kind of resistance. The process of decision making, the Sage and the resistance will all be at the very heart of the present work, since the Devil lurks in the details.